R01 Success (or lack thereof) by K08/K23 Awardees

The Annals of Internal Medicine has a report entitled Sex Differences in Attainment of Independent Funding by Career Development Awardees … but I found the overall attainment statistics quite telling. The authors measured the rates at which recipients of K08 and K23 awards (2,784 awardees from 1997-2003 were studied) subsequently secured R01 awards. The bottom line:

Overall, 31.4% of the 1919 K08 awardees and 43.7% of the 865 K23 awardees were female (P < 0.001). Women were less likely than men to receive an R01 award (P < 0.001). The actuarial rate of R01 award attainment at 5 years was 22.7% overall, 18.8% among women, and 24.8% among men. At 10 years, the rate was 42.5% overall, 36.2% among women, and 45.6% among men. Sex persisted as an independent significant predictor of R01 award attainment (hazard ratio, 0.79 [95% CI, 0.68 to 0.92]; P = 0.002) in multivariate analysis controlling for K award type, year of award, funding institute, institution, and specialty.

This is through the 2007 cycle. Imagine the stats in a couple of years.

The authors were looking at a potential blockage in the pipeline to increase the number women faculty and chairs, noting that “in 2007, women constituted 49% of the medical student body but only 33% of medical faculty, 17% of full professors, and 12% of department chairs at U.S. medical schools.” A couple of their discussion point apply to any K awardee though:

K awards are designed to provide both protected time and mentoring to support the research career development of recipients. However, qualitative evidence collected by the NIH (18) suggests that the financial support offered by K awards may sometimes be insufficient to protect three quarters of the recipient’s working time, as these awards are generally intended to do. Some recipients may face pressure to allocate substantial time to clinical activities. To the extent that women receive smaller awards, protecting time for research may be particularly problematic for them.

The quality of the mentoring relationships of K award recipients also merits further investigation and targeted support. Mentoring is essential to the success of junior investigators in general and may be particularly important for female junior investigators (21–24). Mentoring has long been heralded as a mechanism by which to combat sex disparities in the professions. Some studies have suggested that the quality of mentoring received by women may be inferior to that received by men (15, 25), but others have found no differences (26, 27). K awards require recipients to receive designated mentorship. The low rate of R01 award attainment demonstrated in our study raises important concerns about the quality of mentoring within the K award program and whether mentors are prepared to deal with the special challenges female award recipients face.

The authors acknowledged their time frame may have been too short … but, aside from career breaks for family reasons, I don’t know of too many tenure clocks that extend beyond 10 years, even for clinicians, and the idea is for K awardees to secure their R01 during the K support period. It only gets harder once the K runs out.

Success of career development award recipients, who have a demonstrated aptitude and commitment to research and in whom considerable societal resources have been invested, is critical evidence of whether the physician-scientist pipeline is functioning adequately.

Indeed.

19 Comments »

  1. whimple said

    The sex differences are a distraction from the main point, which seems to be that physician-scientists aren’t very successful as scientists and therefore that the resources put into these K awards are largely (but not completely) wasted. There are at least two non-exclusive possibilities: 1) the physician-scientists tend to be bad scientists. 2) the type of science advocated by the physician-scientists tends to not be appreciated by the largely non-physician-scientist study section population. My guess is that it’s a bit of both. The NIH can fix 2) by forming study sections that are geared towards the type of middle-ground (basic vs clinical) translational science that was the whole impetus behind physician-scientist programs in the first place. 1) can be fixed (as mentioned) by better mentoring.

  2. I, too, wonder about the possibility that K awardees are a selected population of not-so-good scientists.

  3. Neuro-conservative said

    Couldn’t this question be empirically tested by comparing K08/K23 outcomes with K01 awardees? The article doesn’t seem to mention that category.

  4. BB said

    One more thought: you are assuming that the awardee actually gets the protected time.

    Take this scenario (true story): someone got an award but department chair didn’t give awardee protected time. Awardee spent same amount of time in OR, with a part-time lab assistant carrying out experiments. Yes, dept chair (mentor) told granting agency awardee would get x % protected time, but in reality, dept chair kept awardee in OR because patient revenue generates more $$. Could this be happening elsewhere, with more pressure applied to female MDs to see patients?

  5. microfool said

    See NIH’s similar data on the subject on slides 7 and 8 of the Biomedical Research Workforce Report from 2007, which is based on actual grants data, though includes no inferential statistical analysis.

    My read from that data on the post-K08 long-term success is that there is a steady state ~20% of K08 awardees that are never heard from again, and a steady state ~20% of K08 awardees that apply and never get anything, and the rest (~50%) eventually get something. I’d also like to see some long term R01 success rates for the 1980s cohorts.

    NIMH has also has a lengthy report that compares NIMH K01, K08, and K23 awardees from 1999, and their eventual success in getting grants, but again, not restricted to R01-equivalents. K01, K08, and K23 recipients appeared to perform similarly in applying and being awarded subsequent grants thru 2008, which would support CPPs lumping all K awardees together.

    Some you may have luck submitting a data or report request as described at http://grants.nih.gov/grants/funding/welcomewagon.htm.

    Extramural award data is available from NIH’s Research Portfolio Online Reporting Tool (RePORT). RePORT includes a variety of frequently requested reports including a link to the RePORT Expenditures and Results (RePORTER) site where expanded information on funded scientific programs is available. If you have questions about statistics and award data, contact DISHelp@mail.nih.gov.

    Might be easier that brute forcing Google searches.

    • microfool said

      Note, see Figure 11 in the NIMH report for the relevant data.

    • Ryder7 said

      According to the pp , no K08ers recd R01 in 2006 !?!??!!

  6. writedit said

    NIAMS has reported a detailed analysis of the outcome of their training programs, including “subsequent rates of publications and grant support for trainees supported in the past by different types of training grants” (T32, F32, K01, K08). In their cohort:

    17% of T32-supported trainees (n=271 from 93-94);
    34% of F32 awardees (n=44 from 93-94);
    83% of K01 awardees (n=6 from 96-97); and
    55% of K08 awardees (n=58 from 96-97)
    subsequently received an R01;

    37% of T32 trainees, 59% of F32 awardees, and 69% of K08 awardees received other NIH funding.

    These cohorts did better in getting papers out in the past 6 years: 55%, 68% , 100% , and 84%, respectively.

    Lots more interesting data in the report.

    • Delenn said

      RE: BB’s post (10:25 am) and Writedit’s follow up (2:20 pm);

      I find it interesting that NIAMS left out K23s (MENTORED PATIENT-ORIENTED RESEARCH CAREER DEVELOPMENT AWARD) out of their analysis.

      NIAMS is one of the very few NIH entities that have reduced the typical 75% commitment to research down to a 50% commitment if the awardee is a surgeon.

      That said, I can assure you that it is not unusual for surgeons and other physicians to be routinely pressured (either directly or indirectly) to keep their revenue stream flowing, if not increasing, each and every year of their K award, even if their chairs (and other University/Hospital powers-that-be) promised NIH that ‘protected time for research’ would be provided. The creed that University-based Hospital systems must not turn away non-insured or non-paying patients puts enormous pressure on younger/newer faculty as they build both their clinical practice and research programs. I can cite at least four separate examples from four separate institutions where chairs treat only insured patients (a dirty little secret between the appointment schedulers and the chair), while their junior faculty treat the brunt of patients, from which little no revenue will ever be collected. (e.g., Medicare/Medicaid simply don’t reimbure true costs in many if not most cases). Simply put, it’s not an easy environment to do clinical or translational research.

      Bottom line, there are great ideas proposed by great minds; there are also lousy ideas proposed by mediocre scientists, MD or PHD etc. However, with respect to K awards and other NIH funding mechanisms, the playing field between the MD and the PhD, in my opinion, is never equal.

  7. Physician Scientist said

    The true comparison is K08 awardees versus F32 awardees. Both are funded at the early training stage level. Both should have comparable scientific talent. What is the outcome of ultimate R01 funding? The NIAMS data suggests that K08 awardees do better.

    The K01 v. K08 awardees is not necessarily a fair comparison because the K01 is so hard to get that the awardees are generally elite (which is not necessarily true of the K08 awardees) – eg on the level of a Burroughs Wellcome career award.

    In all, I think the data actually suggests that the K08 awardees are doing quite well (esp if they have to also manage clinical work at an early stage of their independent lab life).

    • pinus said

      F32 = K08?

      not at all. several important differences:

      1)An F32 just covers salary with minimal travel/supply monies.

      2)You don’t get to take an F32 with you..when you leave, it ends, not so with K08’s.

      3) K08’s can last up to 5 years (correct me if I am wrong) whereas an F32 is 3 max.

      I am curious how you came to the conclusion that K08 and F32 have simliar talent? Is it because the success rate is similar or some other reason?

      • Physician Scientist said

        I don’t mean to imply that they are equivalent. Clearly they are not. However, in terms of comparison between K08 recipients (MDs and MD/PhDs), I think the F32 is probably the most applicable. Both are a first “independent” grant. Both cover a good portion of salaray (and little else – the K08 supply money is not enough to even cover a tech). Both the F32 recipient and the K08 recipient should be of similar talent in terms of scientific training and are generally at similar stages of their careers – my K08 was funded a year into my post-doc, I assume the F32 is funded at a similar point.

        Although its certainly not a perfect comparison for the reasons you outline above, in order to say that the K08 is not a success, it needs to be compared to something. I think that of the grants on the list the F32 v. K08 comparison is best. That’s all I’m saying.

      • pinus said

        I understand your point, but I still don’t think it is a level comparison, I think the K01 is the suitable data point.

        A K08 provide $75K for salary and $20K for ‘supplies’. An NRSA provides salary, around $38K in year 1 and then $7850 for training related expenses (read insurance instead of that, because that is what most institutions do..make your pay your insurance out of that…not sure about K08’s though). The real kicker is the time and movability. you can take a K08 with you..that is REAL salary support for a position, people like that. A F32 is worth exactly jack shit when trying to move.

  8. John M said

    My observation is that many F32 recipients are actually more experienced scientists at the start of the F32 than MDs starting K08s. MD/PhDs with K08s are farther along. Most PhD scientists applying for an F32 have several papers over 4-6 years of 100% research time. In contrast, the MD K08 awardee often has just 2-3 years of research experience and likely just a few papers. In some specialties this research time is interrupted by quite a bit of clinical time. Thus, at least 5 years of K-supported training period is likely needed–maybe more to finally be independent enough to get an R01. The data showing low conversion to R01 at 5 years but better by 10 yrs supports this. I am an MD physician scientist that only got into basic research during my subspecialty training. I received a K08 after about 2.5 years of research training–of which 10+ weeks/year were spent in patient care. I just received a fundable score on my A1 R01 application (should have started Dec 1, but delayed along with everyone elses grant). This will be 6.5 years after starting my K08. Yes, it takes awhile. Was it worth the government investment to give me a K08 in the first place? We’ll see.

    I agree with whimples first comment. Better mentoring and more appropriate review processes may be needed.

  9. neurowoman said

    Actually, pinus, I had no problem moving my individual NRSA F32. Had to come up with a new mentor & plan and ok it with the P.O., but it made scientific sense.

    The NIMH report (Investing in the Future) seems to say that F32’s only apply for further funding at a 50% rate (receive at 37% rate; Fig. 5); while the K’s apply at 80-100% rates (funded at 60-70%rate; Fig. 11). Similar success rates give the application rate, but I would read the lower application rates for F32s as due to the loss of postdoc’s out of the academic science path (to industry; teaching positions) due to the narrow bottleneck to academic PI-dom. Whereas physicians with K’s have their clinical positions for the indefinite future. Not sure it’s really useful to compare, since the career paths are so different.

    • pinus said

      You moved a F32 to a tenure-track position? I did not know that was possible. I stand corrected.

      • Ryder7 said

        I didn’t know either .

  10. GI guy said

    How many people leave K08 position because of large disparity between compensation between that and private practice? Has K08 award kept pace with growth in private compensation? Does anyone have data on this? This is clearly a disincentive for physician-scientists who want to pursue research career but have financial obligations/debt burden due to long clinical training. It seems counterproductive to encourage people to enter this training process and then not support them in the vulnerable junior faculty stage with at least a somewhat comparable salary (academic salaries are always 70% of private, but K08 is clearly below this rate).

  11. GI fellow said

    One striking thing is that only around 10% KO8 or K23 awardees from 2003 received RO1. I have seen too many junior faculties give up research and going for private or becoming a full time clinician.

RSS feed for comments on this post · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: